Architecture Quality Revisited

Abstract

A recent study suggests that nonfunctional quality is of little relevance to users and customers but mainly a concern for architects. Nontechnical constraints appear to be driving design as prominently as quality requirements. The PragmaTic archiTecTA recent study, "Non-Functional Requirements in Software Architecture Practice," investigates how architects deal with nonfunctional requirements (NFRs) in their daily practices. 1 The results appear contradictory to the common belief that nonfunctional quality is fundamentally important for architecture sustainability and project success. 2,3 In a nutshell, the study suggests that nonfunctional quality is of little relevance for users and customers, and is instead primarily a concern for software architects. It's thus tempting to conclude that practitioners consider nonfunctional qualities as an afterthought, rather than as a prime driver of architecture design. The study raises several thoughtprovoking questions: Does the research community actually overestimate the relevance of nonfunctional qualities? Is practitioners' understanding about the real importance of architecture quality amiss? What do these results imply? the studyThe study focused on 13 software architects working in Spanish organizations covering a spectrum of application domains. It used exploratory qualitative research with semistructured interviews centered on a single project. The projects themselves varied in functionality, size, and project staff involvement.In this column, we focus on questions and results from the study that we believe have the biggest impact on actual architecture design:

Prominent quality attributes of crisis software systems: a literature review

“…In [12], software quality attributes were presented within two categories that are executional and evolutionary for detailed quality attribute definitions [24]. Table 2 shows executional (operational) quality attributes and the impact of these quality attributes observed in action, during which users actively perform their tasks [25]. Table 3 presents evolutionary (developmental) quality attributes defined as "capability of a set of static attributes of a software product to satisfy the stated and implied needs when the software product is used under specified conditions" [26,27].…”

Non-functional requirements as drivers of software architecture design

“…We may cite as clear example in our study the leading role of software architects as the main source of NFRs. For this particular issue, a reflection on the variability and importance of the software architect role depending on the type of project can be found at [50]. As a beneficial side-effect, after an hour or more of being interviewed, respondents learn about certain issues to which they are not usually exposed (e.g., NFR terminology) and some of them expressed their desire to collaborate again on future studies (as it has really happened).…”

“…In response to this study we collaborated with Frank Buschmann in writing a column in the IEEE Software magazine [50] highlighting some of the most relevant outcomes of the study. After these two last publications we got accepted a research paper to the IEEE Software magazine but this time focusing on the architectural decisions [12].…”

“…: 1.508 3 [16] NFRs into MDD DSDM Workshop 2 [14] Architectural Knowledge RE CORE A 2 [17] NFRs into MDD Euromicro CORE C 2 [13] Architectural Knowledge TOPI Workshop 1 [15] Architectural Knowledge SPE I.F. : 0.519 1 [19] Architectural Knowledge DSDM Workshop 1 [20] Empirical research EASA Workshop 0 [50] Empirical research IEEE Soft. I.F.…”

“…• Second empirical study. This empirical study was designed in 2009, executed in 2009-2010, and the results were published in 2012 [11,50,12]. The study is based on 13 personal interviews to software architects.…”